[EM] 28 years of progress and a wakeup call

[Closed Limelike Curves]

There's some difficulty here in that Duverger's law has been used
inconsistently. I'd agree the statement of Duverger's law on Wikipedia is
currently quite bad. I'd also note—like you did—that it was bad even before
I edited it, and I think this is an improvement: going from "every
single-winner system leads to two-party rule" to "single-winner systems
that don't comply with FBC lead to two-party rule" is *less* incorrect,
even if it's still incorrect, because it no longer (incorrectly) states
that FBC-compliant single-winner systems lead to a 2-party system. I've
just provided an additional citation to an author using FBC in this sense.

Duverger's original paper, like you say, makes the claim that FPP
inevitably leads to a two-party system, whereas top-2 runoff and
proportional representation create multi-party systems. Political
scientists have abandoned this framing because it's simply wrong: many
top-2 runoff countries have two-party systems, and quite a few FPP
countries *don't* have two-party systems. Top-2 runoff has a *weaker* push
towards two-party rule than FPP, in large part because top-2 fails sincere
favorite less often. But 2RS countries still have noticeably fewer parties
on average than proportional representation countries, and game theory
models of strategic voting tend to find that 2RS countries have stable
equilibria at 2 or 3 party systems.

I'd love it if the article were rewritten to reflect all this nuance, but
there are a finite number of hours in the day for me.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20240526/ceabd275/attachment.htm>